OBJECTIVE: To investigate the effect of gestational weight gain in obese glucose-tolerant women. RESEARCH DESIGN AND METHODS: We performed a historical cohort study of 481 women with prepregnancy BMI > or = 30 kg/m2 and a normal 2-h 75-g oral glucose tolerance test (OGTT) during the third trimester (according to World Health Organization criteria). Data on OGTT results and clinical outcomes were collected from medical records. Four groups were defined according to weight gain: group 1, <5.0 kg (n = 93); group 2, 5.0-9.9 kg (n = 134); group 3, 10.0-14.9 kg (n = 132); and group 4, > or = 15.0 kg (n = 122). RESULTS: Birth weight increased significantly with increasing weight gain (mean grams +/- SD): group 1, 3,456 +/- 620; group 2, 3,624 +/- 675; group 3, 3,757 +/- 582; and group 4, 3,784 +/- 597 (P < 0.001). The birth weight in group 1 was similar to that of the background population of primarily normal-weight women (3,478 g). In multivariate analyses, increasing weight gain was associated with significantly higher rates of hypertension (OR 4.8 [95% CI for group 4 vs. group 1: 1.7-13.1]), cesarean section (3.5 [1.6-7.8]), induction of labor (3.7 [1.7-8.0]), and large-for-gestational-age infants (4.7 [2.0-11.0]). There was no difference in rates of small-for-gestational-age infants. Significant predictors for birth weight (determined by multiple linear regression) were gestational weight gain, 2-h OGTT result, pre-gestational BMI, maternal age, gestational age, and smoking. CONCLUSIONS: Increasing weight gain in obese women is associated with increasing pregnancy complications. Our data suggest that minimal gestational weight gain might normalize birth weight. Prospective studies should be performed to clarify the safety of recommending limited gestational weight gain.